降质图像复原方法研究
|
摘要
图像作为人类感知客观世界及事物存在的重要工具,不仅能够形象地表达物理对象,而且能够使人类更加直观地认知并理解物理对象间的关系,进而使人类能够更加全面、深刻地研究并探索物理世界。在人类通过各种手段及不同观测系统从客观世界获取图像的过程中,存在各种干扰因素(如各类噪声、运动模糊、大气湍流模糊、散焦模糊等),使实际获取图像的质量下降,进而令其失去实际应用价值。因此,复原图像原始信息并提高其图像质量进而还原真实场景具有非常重要的意义。
本文针对噪声干扰、运动模糊、大气湍流模糊及散焦模糊产生的降质图像复原问题,分别构建降质图像复原模型并设计相应的复原算法,以去除图像降质现象、提高图像质量与保护图像细节信息等因素为切入点研究降质图像复原的新方法,其主要内容及创新点如下：
(1)针对图像在获取、记录、转换及传输过程中易受到噪声污染的问题,提出了一种新的基于Chebyshev理论与Radon变换的噪声降质图像复原方法。该方法基于Chebyshev理论和Radon变换构建新的自适应参数辨识去噪模型,并设计与该模型相匹配的新算法,旨在去除脉冲噪声、高斯噪声及其混合噪声(包括脉冲噪声、高斯噪声等多类噪声混合),通过噪声检测和噪声去除两个阶段完成噪声降质图像的复原工作。经实验仿真,以图像质量主观和客观评价为参考,将该方法与常见的方法比较,实验结果表明文中所提出的方法在降质图像的去噪能力和图像细节保护能力方面均具有较好的性能。
(2)针对图像在摄取过程中因被摄对像与成像设备间相对运动引发运动模糊降质问题,提出了一种新的基于反应扩散理论的运动模糊降质图像复原方法。该方法利用反应扩散方程的图像复原与图像边缘信息保护的特征,结合运动模糊图像量化函数,基于反应扩散理论构建新的运动模糊降质图像复原模型,并设计与该模型相匹配的新算法,旨在去除图像的运动模糊,完成该类降质图像的复原工作。本文给出了基于反应扩散理论的运动模糊降质图像复原模型及其相关证明,并经实验仿真,以图像质量主观和客观评价为参考,将该方法与常见的方法比较,实验结果表明文中所提出的方法在降质图像的去除运动模糊能力和图像细节保护能力方面均具有较好的性能,并对噪声干扰具有一定的鲁棒性。
(3)针对图像在获取过程中易受到大气湍流影响引发大气湍流模糊降质问题,提出了-种新的基于图像相似度-反应扩散理论的大气湍流模糊降质图像复原方法。该方法利用图像相似度的特征与反应扩散理论的特性,结合光学相似度函数,基于图像相似度-反应扩散理论构建新的大气湍流模糊降质图像复原模型,并设计与该模型相匹配的新算法,旨在去除图像的大气湍流模糊,完成该类降质图像的复原工作。经实验仿真,选取遥感图像并以图像质量主观和客观评价为参考,将该方法与常见的方法比较,实验结果表明文中所提出的方法在降质图像的去除大气湍流模糊能力和图像细节保护能力方面均具有较好的性能,并对噪声干扰具有一定的鲁棒性。
(4)针对图像在获取过程中普遍存在因聚焦不准引发散焦模糊问题,提出了一种新的基于结构张量-反应扩散理论的散焦模糊降质图像复原方法。该方法利用结构张量可分析并处理图像中结构信息的方向场、边缘、角点及其它几何特征的属性与反应扩散理论的特性,结合扩散张量模型,基于结构张量-反应扩散理论构建新的散焦模糊降质图像复原模型,并设计与该模型相匹配的新算法,旨在去除图像的散焦模糊,完成该类降质图像的复原工作。经实验仿真,以图像质量主观和客观评价为参考,将该方法与常见的方法比较,实验结果表明文中所提出的方法在降质图像的去除散焦模糊能力和图像细节保护能力方面均具有较好的性能,并对噪声干扰具有一定的鲁棒性。
(5)基于图像降质因素分析与图像质量评价,以针对各类常见图像降质现象设计并实现降质图像复原模型与算法为基础工作,提出了一种降质图像复原系统的原型设计。该系统原型设计旨在结合降质图像复原方法研究中已取得的成果给出一种简单、高效、适用范围较广且易于实现的降质图像复原解决方案。该方案以满足现有图像处理需求并解决当前图像复原问题为目标,便于将降质图像复原方法转化为产品并应用于日常生活各方面。
本文专注于研究降质图像复原问题,描述了图像降质成因、降质分类及其数学模型,针对噪声干扰、运动模糊、大气湍流模糊、散焦模糊等主要图像降质模型,设计了适用不同降质现象的图像复原算法,以提高复原图像质量与保护图像细节信息等因素为参考不断提升算法性能,最终给出了一种简单、高效、适用范围较广且易于实现的降质图像复原解决方案,为图像处理领域中降质图像复原技术的研究提供了新方法与新思路。
Image is an important tool for human perceiving the existence of the world, not only can express the physical objects but also can make people understand the relationship between the recognizable objects much more conveniently, so that human could more profoundly study the physical world. The images can be acquired from the objective world by different observation systems and various means, however, there are all kinds of adverse factors during the image capture, such as various noises, motion blurring, atmospheric turbulence blurring, defocus blurring and so on. The images can be degraded, and then lose the worthy of application for real environment. Therefore, it is the very important value that the original scene can be restored and the quality the images can be improved.
In this thesis, we have established several degraded image restoration models and designed the related algorithms to filter noise, motion blurring, atmospheric turbulence blurring and defocus blurring from the degraded images, and have studied the performance of the methods in terms of the image restoration quality and image detail preservation. The main content and innovation points are listed as follows:
(1) To address the problem of different types of noise, by which the images are corrupted when they are captured, stored, converted and transmited, a new method based on Chebyshev theory and Radon transform is proposed for filtering the single noise or mixed noise from the corrupted images, PA for short. This method PA, where the parameters are identified adaptively in the related model, contains two stages:noise detection and noise reduction, which would remove impulse noise, Gaussian noise and other various noises mixed by impulse noise and Gaussian noise in different proportions. Extensive experimental results show that the proposed method PA outperforms the other classical methods in terms of subjective and objective image quality assessment, and demonstrate that the proposed model and algorithm PA is an effective image restoration method in the image denoising and image detail preservation.
(2) To address the problem of motion blurring in the process of taking a image, which is mainly caused by the relative motion between the subject and the imaging equipment, a new method based on reaction-diffusion equation is proposed for restoring images that are degraded by motion blurring, RDER for short. This method RDER focuses on constructing the image restoration model and designing the related algorithm for filtering motion blurring, which takes advantage of motion blurring image quantization function and characteristics of reaction-diffusion equation about restoring image and preserve edge information. In particular, we introduce the image restoration capability of the RDER by model analysis and mathematical theory. To demonstrate the advantages of the RDER, which is compared with some common deblurring methods, extensive experimental results show that the proposed method RDER outperforms the other classical methods in terms of subjective and objective image quality assessment, and demonstrate that the proposed model and algorithm RDER is an effective image restoration method in the image motion deblurring and image detail preservation, meanwhile, has a certain robustness to noise interference.
(3) To address the problem of image blurring caused by the atmospheric turbulence effect, a new method based on the image similarity and reaction-diffusion equation is proposed for restoring images that are degraded by the atmospheric turbulence blurring, ISRDE for short. The main task of this method ISRDE is to construct the image restoration model and design the related algorithm for filtering atmospheric turbulence blurring, which takes advantage of characteristics of optical similarity function and reaction-diffusion equation (RDE) about restoring image. In particular, several real remote sensing images are used in comparative experiment of algorithm to improve the application value of algorithm. Extensive experimental results show that the proposed method ISRDE outperforms the other classical methods in terms of subjective and objective image quality assessment, and demonstrate that the proposed model and algorithm ISRDE is an effective image restoration method in the image atmospheric turbulence deblurring and image detail preservation, meanwhile, has a certain robustness to noise interference.
(4) To address the problem of image blurring caused by defocus, a new method based on the structure tensor and reaction-diffusion equation is proposed for restoring images that are degraded by the defocus blurring, STRDE for short. The main task of this method STRDE is to construct the image restoration model and design the related algorithm for removing defocus blurring, which takes full advantage of the geometric characteristics of the structure tensor, which can analys and process the structure information (for example, orientation field, edge, corner and so on), and reaction- diffusion equation (RDE) about restoring image. Extensive experimental results show that the proposed method STRDE outperforms the other classical methods in terms of subjective and objective image quality assessment, and demonstrate that the proposed model and algorithm STRDE is an effective image restoration method in the image defocus deblurring and image detail preservation, meanwhile, has a certain robustness to noise interference.
(5) To address the problem of the degraded image restoration, combining the proposed methods of image denoising, motion deblurring, atmospheric turbulence deblurring and defocus deblurring, we have given the design of a degraded image restoration prototype system, which describes structure diagrams and general workflows of the prototype system based on the mathematical models, related algorithms and key technology. It is a simple and efficient solution, which is ease of implementation and has a wide range of application which is used in our daily life from military to civilian, from industry to commerce and so on.
This thesis focuses on the research on degraded image restoration, where we have described the main factors causing image degradation, the categories of degraded factors and their mathematical models including various image noises, motion blurring, atmospheric turbulence blurring and defocus blurring. We have designed different degraded image restoration models and algorithms for removing different image degraded factors and improved the performance of these models in terms of the image restoration quality and image detail preservation. Finally, we have given an integrated solution for degraded image restoration, which is applied widely and translated into the product easily. In conclusion, the research works of this thesis give a new and useful attempt for degraded image restoration and will provide support and guarantee to the other researches in image processing.
本文针对噪声干扰、运动模糊、大气湍流模糊及散焦模糊产生的降质图像复原问题,分别构建降质图像复原模型并设计相应的复原算法,以去除图像降质现象、提高图像质量与保护图像细节信息等因素为切入点研究降质图像复原的新方法,其主要内容及创新点如下：
(1)针对图像在获取、记录、转换及传输过程中易受到噪声污染的问题,提出了一种新的基于Chebyshev理论与Radon变换的噪声降质图像复原方法。该方法基于Chebyshev理论和Radon变换构建新的自适应参数辨识去噪模型,并设计与该模型相匹配的新算法,旨在去除脉冲噪声、高斯噪声及其混合噪声(包括脉冲噪声、高斯噪声等多类噪声混合),通过噪声检测和噪声去除两个阶段完成噪声降质图像的复原工作。经实验仿真,以图像质量主观和客观评价为参考,将该方法与常见的方法比较,实验结果表明文中所提出的方法在降质图像的去噪能力和图像细节保护能力方面均具有较好的性能。
(2)针对图像在摄取过程中因被摄对像与成像设备间相对运动引发运动模糊降质问题,提出了一种新的基于反应扩散理论的运动模糊降质图像复原方法。该方法利用反应扩散方程的图像复原与图像边缘信息保护的特征,结合运动模糊图像量化函数,基于反应扩散理论构建新的运动模糊降质图像复原模型,并设计与该模型相匹配的新算法,旨在去除图像的运动模糊,完成该类降质图像的复原工作。本文给出了基于反应扩散理论的运动模糊降质图像复原模型及其相关证明,并经实验仿真,以图像质量主观和客观评价为参考,将该方法与常见的方法比较,实验结果表明文中所提出的方法在降质图像的去除运动模糊能力和图像细节保护能力方面均具有较好的性能,并对噪声干扰具有一定的鲁棒性。
(3)针对图像在获取过程中易受到大气湍流影响引发大气湍流模糊降质问题,提出了-种新的基于图像相似度-反应扩散理论的大气湍流模糊降质图像复原方法。该方法利用图像相似度的特征与反应扩散理论的特性,结合光学相似度函数,基于图像相似度-反应扩散理论构建新的大气湍流模糊降质图像复原模型,并设计与该模型相匹配的新算法,旨在去除图像的大气湍流模糊,完成该类降质图像的复原工作。经实验仿真,选取遥感图像并以图像质量主观和客观评价为参考,将该方法与常见的方法比较,实验结果表明文中所提出的方法在降质图像的去除大气湍流模糊能力和图像细节保护能力方面均具有较好的性能,并对噪声干扰具有一定的鲁棒性。
(4)针对图像在获取过程中普遍存在因聚焦不准引发散焦模糊问题,提出了一种新的基于结构张量-反应扩散理论的散焦模糊降质图像复原方法。该方法利用结构张量可分析并处理图像中结构信息的方向场、边缘、角点及其它几何特征的属性与反应扩散理论的特性,结合扩散张量模型,基于结构张量-反应扩散理论构建新的散焦模糊降质图像复原模型,并设计与该模型相匹配的新算法,旨在去除图像的散焦模糊,完成该类降质图像的复原工作。经实验仿真,以图像质量主观和客观评价为参考,将该方法与常见的方法比较,实验结果表明文中所提出的方法在降质图像的去除散焦模糊能力和图像细节保护能力方面均具有较好的性能,并对噪声干扰具有一定的鲁棒性。
(5)基于图像降质因素分析与图像质量评价,以针对各类常见图像降质现象设计并实现降质图像复原模型与算法为基础工作,提出了一种降质图像复原系统的原型设计。该系统原型设计旨在结合降质图像复原方法研究中已取得的成果给出一种简单、高效、适用范围较广且易于实现的降质图像复原解决方案。该方案以满足现有图像处理需求并解决当前图像复原问题为目标,便于将降质图像复原方法转化为产品并应用于日常生活各方面。
本文专注于研究降质图像复原问题,描述了图像降质成因、降质分类及其数学模型,针对噪声干扰、运动模糊、大气湍流模糊、散焦模糊等主要图像降质模型,设计了适用不同降质现象的图像复原算法,以提高复原图像质量与保护图像细节信息等因素为参考不断提升算法性能,最终给出了一种简单、高效、适用范围较广且易于实现的降质图像复原解决方案,为图像处理领域中降质图像复原技术的研究提供了新方法与新思路。
Image is an important tool for human perceiving the existence of the world, not only can express the physical objects but also can make people understand the relationship between the recognizable objects much more conveniently, so that human could more profoundly study the physical world. The images can be acquired from the objective world by different observation systems and various means, however, there are all kinds of adverse factors during the image capture, such as various noises, motion blurring, atmospheric turbulence blurring, defocus blurring and so on. The images can be degraded, and then lose the worthy of application for real environment. Therefore, it is the very important value that the original scene can be restored and the quality the images can be improved.
In this thesis, we have established several degraded image restoration models and designed the related algorithms to filter noise, motion blurring, atmospheric turbulence blurring and defocus blurring from the degraded images, and have studied the performance of the methods in terms of the image restoration quality and image detail preservation. The main content and innovation points are listed as follows:
(1) To address the problem of different types of noise, by which the images are corrupted when they are captured, stored, converted and transmited, a new method based on Chebyshev theory and Radon transform is proposed for filtering the single noise or mixed noise from the corrupted images, PA for short. This method PA, where the parameters are identified adaptively in the related model, contains two stages:noise detection and noise reduction, which would remove impulse noise, Gaussian noise and other various noises mixed by impulse noise and Gaussian noise in different proportions. Extensive experimental results show that the proposed method PA outperforms the other classical methods in terms of subjective and objective image quality assessment, and demonstrate that the proposed model and algorithm PA is an effective image restoration method in the image denoising and image detail preservation.
(2) To address the problem of motion blurring in the process of taking a image, which is mainly caused by the relative motion between the subject and the imaging equipment, a new method based on reaction-diffusion equation is proposed for restoring images that are degraded by motion blurring, RDER for short. This method RDER focuses on constructing the image restoration model and designing the related algorithm for filtering motion blurring, which takes advantage of motion blurring image quantization function and characteristics of reaction-diffusion equation about restoring image and preserve edge information. In particular, we introduce the image restoration capability of the RDER by model analysis and mathematical theory. To demonstrate the advantages of the RDER, which is compared with some common deblurring methods, extensive experimental results show that the proposed method RDER outperforms the other classical methods in terms of subjective and objective image quality assessment, and demonstrate that the proposed model and algorithm RDER is an effective image restoration method in the image motion deblurring and image detail preservation, meanwhile, has a certain robustness to noise interference.
(3) To address the problem of image blurring caused by the atmospheric turbulence effect, a new method based on the image similarity and reaction-diffusion equation is proposed for restoring images that are degraded by the atmospheric turbulence blurring, ISRDE for short. The main task of this method ISRDE is to construct the image restoration model and design the related algorithm for filtering atmospheric turbulence blurring, which takes advantage of characteristics of optical similarity function and reaction-diffusion equation (RDE) about restoring image. In particular, several real remote sensing images are used in comparative experiment of algorithm to improve the application value of algorithm. Extensive experimental results show that the proposed method ISRDE outperforms the other classical methods in terms of subjective and objective image quality assessment, and demonstrate that the proposed model and algorithm ISRDE is an effective image restoration method in the image atmospheric turbulence deblurring and image detail preservation, meanwhile, has a certain robustness to noise interference.
(4) To address the problem of image blurring caused by defocus, a new method based on the structure tensor and reaction-diffusion equation is proposed for restoring images that are degraded by the defocus blurring, STRDE for short. The main task of this method STRDE is to construct the image restoration model and design the related algorithm for removing defocus blurring, which takes full advantage of the geometric characteristics of the structure tensor, which can analys and process the structure information (for example, orientation field, edge, corner and so on), and reaction- diffusion equation (RDE) about restoring image. Extensive experimental results show that the proposed method STRDE outperforms the other classical methods in terms of subjective and objective image quality assessment, and demonstrate that the proposed model and algorithm STRDE is an effective image restoration method in the image defocus deblurring and image detail preservation, meanwhile, has a certain robustness to noise interference.
(5) To address the problem of the degraded image restoration, combining the proposed methods of image denoising, motion deblurring, atmospheric turbulence deblurring and defocus deblurring, we have given the design of a degraded image restoration prototype system, which describes structure diagrams and general workflows of the prototype system based on the mathematical models, related algorithms and key technology. It is a simple and efficient solution, which is ease of implementation and has a wide range of application which is used in our daily life from military to civilian, from industry to commerce and so on.
This thesis focuses on the research on degraded image restoration, where we have described the main factors causing image degradation, the categories of degraded factors and their mathematical models including various image noises, motion blurring, atmospheric turbulence blurring and defocus blurring. We have designed different degraded image restoration models and algorithms for removing different image degraded factors and improved the performance of these models in terms of the image restoration quality and image detail preservation. Finally, we have given an integrated solution for degraded image restoration, which is applied widely and translated into the product easily. In conclusion, the research works of this thesis give a new and useful attempt for degraded image restoration and will provide support and guarantee to the other researches in image processing.
引文
[1]王耀德,陈家琪,黄文华.论技术创新的起源和动力[M].北京：北京理工大学出版社,2006.
[2]孙即祥.图像处理[M].北京：科学出版社,2004.
[3]G. A. Baxes. Digital Image Processing:Principles and Applications [M]. New York,1994:20-25.
[4]X. Wang, X. Q. Zhao, F. X. Guo, J. F. Ma. Impulsive Noise Detection by Double Noise Detector and Removal using Adaptive Neural-fuzzy Inference System [J]. Int. J. Electron.Commun(AEu),2011,65(5):429-434.
[5]A. K. Katsaggelos, M. R. Schroeder, T. Kohonen, et al. Digital image restora-tion [M]. Springer-Verlag New York Inc.,1991.
[6]T. S. Cho, C. L. Zitnick, N. Joshi, et al. Image Restoration by Matching Gradient Distributions [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(4):683-694.
[7]冈萨雷斯.数字图像处理(第二版)[M].北京：电子工业出版社,2003.
[8]A. Kour, V. K. Yadav, V. Maheshwari. A Review on Image Processing [J]. International Journal of Electronics Communication and Computer Engineering, 2013,4(1):270-275.
[9]M. Egmont-Petersen, D. de Ridder, H. Handels. Image Processing with Neural Networks-A Review [J]. Pattern Recognition,2002,35(10):2279-2301.
[10]J. Mairal, F. R. Bach, J. Ponce, et al. Non-local Sparse Models for Image Restoration [C]. In Proc. of the 2009 IEEE 12th International Conference on Computer Vision, Kyoto, Sept.29-Oct.2,2009:2272-2279.
[11]T. Ch. Lin. Decision-based Fuzzy Image Restoration for Noise Reduction Based on Evidence Theory [J]. Expert Systems with Applications,2011,38(7): 8303-8310.
[12]M. Bertalmio, G. Sapiro, V. Caselles, et al. Image Inpainting [C]. In Proc. of the International Conference on Computer Graphics and Interactive Techniques, Louisiana, USA,2000:417-424.
[13]G. Xu, E. Tang. Blind Deconvolution for Image Restoration Based on Text Characteristic[C]. In Proc. of the Control,2011 International Conference on Automation and Systems Engineering (CASE), Singapore, July 30-31,2011:1-3.
[14]M R Banham, A K Katsaggelos. Digital image restoration [J]. IEEE Signal Pro-cessing Magazine,1997,14(2):24-41.
[15]Computer Vision Research Groups of Carnegie Mellon University [EB/OL]. H-ttp://www.cs.cmu. edu/-cil/vision.html,2013-03-12/2013-03-12.
[16]Computer Vision and Robotics Group of Cambridge University [EB/OL]. http:/ /mi.eng.cam.ac.uk/mi/,2013-03-12/2013-03-12.
[17]Vision and Image Analysis Group of Cornell University [EB/OL]. http://ww-w.via.cornell.edu/,2013-03-12/2013-03-12.
[18]Harvard Robotics Laboratory [EB/OL]. http://hrl.harvard.edu/,2013-03-12/ 2013-03-12.
[19]Zucker Computational Vision Group [EB/OL]. http://cs-www.cs.yale.edu/hom-es/vision/zucker/,2013-03-12/2013-03-12.
[20]Stanford Vision Laboratory [EB/OL]. http://vision.stanford.edu/,2013-03-12/ 2013-03-12.
[21]Biometerics Research Group of Michican State University [EB/OL]. http://w-ww.cse.msu.edu/biometrics/,2013-03-12/2013-03-12.
[22]Computer Vision Group located within the Division Ⅲ of the Computer Science Department in the University of Bonn [EB/OL]. http.//www-dbv.informati-k.uni-bonn.de/,2013-03-12/2013-03-12.
[23]Queen Mary Vision Laboratory [EB/OL]. http://www.dcs.qmul.ac.uk/research/ vision/,2013-03-12/2013-03-12.
[24]Technical Committee on Pattern Analysis and Machine Intelligence (TCPAMI) [EB/OL]. http://www.computer.org/portal/web/tcpami,2013-03-12/2013-03-12.
[25]IPDL世界知识产权组织电子图书馆[EB/OL]. http://www.wipo.int/wipogold/ en/,2013-03-01/2013-03-01.
[26]中国科学院沈阳自动化研究所机器人视觉研究组[EB/OL]. http://vision. sia.cn/visiongroup%28chinese%29.html,2013-03-12/2013-03-12.
[27]华中科技大学图像识别与人工智能研究所[EB/OL]. http://iprai.hust.edu.cn/, 2013-03-12/2013-03-12.
[28]东南大学影像科学与技术实验室[EB/OL]. http://www.imagetech.com.cn/, 2013-03-12/2013-03-12.
[29]浙江大学应用数学与图像处理研究所[EB/OL]. http://www.cms.zju.edu.cn/ frontindex.asp,2013-03-12/2013-03-12.
[30]国家自然科学基金委员会资助项目统计[EB/OL]. http://www.nsfc.gov. cn/nsfc/cen/xmtj/index.html,2013-03-12/2013-03-12.
[31]沈垣,李舜酩,毛建国,辛江慧.数字图像复原技术综述[J].中国图象图形学报,2009,14(9)：1764-1775.
[32]E. O. Brigham, H. W. Smith, F. X. Bostick, et al. An Iterative Technique for Determining Inverse Filters [J]. IEEE Trans, on Geoscience Electronics,1968, 6(2):86-96.
[33]N. Mastronardi, P. Lemmerling, A. Kalsi, et al. Implementation of the Regularized Structured Total Least Squares Algorithms for Blind Image Deblurring [J]. Linear Algebra and its Applications,2004,391(1):203-221.
[34]B. Chalmond. Image Restoration Using an Estimated Markov Model [J]. Sign-al Processing 1988,15(2):115-129.
[35]M. I. Sezan, A. M.T ekalp. Adaptive Nonlinear Image Restoration Using the Theory of Convex Projections [J]. Circuits and Systems,1988,7(9):959-962.
[36]Jr. N. C. Gallagher, G. L. Wise. A Theoretical Analysis of Properties of the Median Filters [J]. IEEE Trans. On Acoustics Speech, Signal Processing,1981, 29(1):1136-1141.
[37]J. Serra. Morphological Filtering:An overview [J]. Signal Processing,1994,38 (1):3-11.
[38]F. Farbiz, M. Menhaj, S. Motamedi, and M. Hagan. A New Fuzzy Logic Filter for Image Enhancement [J]. IEEE Trans. Syst, Man, Cybern. B, Cybern.2000, 30(1):110-119.
[39]P. Windyga. Fast Impulsive Noise Removal [J]. IEEE Transaction on Image Process.2001,10(1):173-179.
[40]Y. T. T. Zhou, R. Chellappa, A. Vaid. Image Restoration Using a Neural Net-work [J]. IEEE Transactions on Acoustics, Speech and Signal Processing,1988, 36(7):1141-1151.
[41]Y. D. Wu, Y. H. Chen, H. Y. Zhang. An Improved Algorithm for Image Restora-tion based on Modified Hopfield Neural Network [J]. Machine Learning and Cybernetics,2005,48(7):4720-4723.
[42]L. J. Deng, T. Zh. Huang, X. L. Zhao. Wavelet-based Two-level Methods for Image Restoration [J]. Communications in Nonlinear Science and Numerical Simulation,2012,17(12):5079-5087.
[43]L. P. Zhang, H. Y. Zhang, H. F. Shen, P. X. Li. A Super-resolution Reconstruction Algorithm for Surveillance Images [J]. Signal Processing,2010,90(3):848-859.
[44]Ch. P. Sung, K. P. Min, G K. Moon. Super-resolution Image Reconstruction:a Technical Overview [J]. IEEE Signal Processing Magazine,2003,20(3):21-36.
[45]X. F. Wan, Y. Yang. Super-resolution Image Reconstruction[C]. In Proc. of the 2010 International Conference on Computer Application and System Modeling (ICCASM), Taiyuan, Oct.22-24,2010(8):351-355.
[46]K. K. Delibasis, P. E. Undrill, G G Cameron. Genetic Algorithm Implement-ation of Stack Filter Design for Image Restoration [J]. IEE Proceedings Vision, Image and Signal Processing,1996,143(3):177-183.
[47]E. Y. Kim, K. T. Kim, B. Kim. Genetic Algorithm-based Reconstruction of Old Films Corrupted by Scratches and Blotches [J]. Pattern Recognition Letters,2013, 34(2):226-237.
[48]O. P. Verma, P. Kumar, M. Hanmandlu, S. Chhabra. High Dynamic Range Optimal Fuzzy Color Image Enhancement Using Artificial Ant Colony System [J]. Applied Soft Computing,2012,12(1):394-404.
[49]A. M. A. C. Rocha, M. F. P. Costa. Artificial Fish Swarm Filter-Based Method for Constrained Global Optimization [J]. Computational Science and Its Appli-cations-ICCSA,2012,7335:57-71.
[50]A. Buades, B. Coll, and J. M. Morel. A Review of Image Denoising Algorithms, with a New One [J]. Multiscale Model Simul.,2006,4(2),490-530.
[51]徐树奎.基于计算摄影的运动模糊图像清晰化技术研究[D].湖南：国防科技大学,2011.
[52]J. F. Cai, B. Dong, S. Osher, Z. W. Shen. Image Restoration:Total Variation, W-avelet Frames and Beyond [J]. J. Amer. Math. Soc,2012,25:1033-1089.
[53]S. Zheng, Z. Pan, G Wang, X. Y. Audio. A Variational Model of Image Restoration Based on First and Second Order Derivatives and Its Split Bregman algorithm[C]. In Proc. of the 2012 International Conference on Audio, Language and Image Processing (ICALIP), Shanghai, July 16-18,2012:860-865.
[54]C. Liu, W. T. Freeman, R. S. Szeliski, et al. Noise Estimation from a Single Im-age [C]. In Proc. of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 17-22,2006,1:901-908.
[55]Y. Hwang J. S. Kim, I. S. Kweon. Difference-Based Image Noise Modeling Us-ing Skellam Distribution [J]. IEEE Transaction on Pattern Analysis and Machi-ne Intelligence,2012,34(7):1329-1341.
[56]Y. Y. Zhou, Zh. F. Ye, Y. Xiao. A Restoration Algorithm for Images Contamina-ted by Mixed Gaussian Plus Random-valued Impulse Noise [J]. Journal of Vis-ual Communication and Image Representation,2013,24(3):283-294.
[57]Zh. P. Wei, J. Wang, H. Nichol, S. Wiebe, D. Chapman. A Median-Gaussian Fi-ltering Framework for Moire Pattern Noise Removal from X-ray Microscopy I-mage [J]. Micron,2012,43(2-3):170-176.
[58]Ch. T. Lu, T. Ch. Chou. De-noising of Salt-and-pepper Noise Corrupted Image Using Modified Directional-weighted-median Filter [J]. Pattern Recognition L-etters,2012,33(10):1287-1295.
[59]Y. Han, X. Ch. Feng, G. Baciu, W. W. Wang. Nonconvex Sparse Regularizer B-ased Speckle Noise Removal [J]. Pattern Recognition,2013,46(3):989-1001.
[60]邹谋炎.反卷积和信号复原[M].北京：国防工业出版社,2001.
[61]刘益民.图像运动模糊机理与恢复技术方法的研究[D].南京：南京理工大学,2008.
[62]刘明.基于图像复原的航空摄影前向像移检测及补偿技术研究[D].长春：中国科学院研究生院(长春光学精密机械与物理研究所),2005.
[63]M. Yang, Ch. L. Gong. Underwater Image Restoration by Turbulence Model Based on Image Gradient Distribution[C]. In Proc. of the 2nd International Co-nference on Uncertainty Reasoning and Knowledge Engineering (URKE), Jal-arta, Aug 14-15,2012:296-299.
[64]李庆菲.大气湍流退化图像复原研究[D].安徽：合肥工业大学,2010.
[65]D. F. Shi, Ch. Y. Fan, H. Shen, P. F. Zhang, et al. Restoration of Atmospheric T-urbulence Degraded Images [J]. Optics Communications,2011,284(24):5556-5561.
[66]H. Y. Hong, L. Ch. Li, T. X. Zhang. Blind Restoration of Real Turbulence-degr aded Image with Complicated Backgrounds Using Anisotropic Regularization [J]. Optics Communications,2012,285(24):4977-4986.
[67]V. P. Aksenov, I. V. Izmailov, F. Y. Kanev, B. N. Poizner. Optical Vortex Detect-or as A Basis for A Data Transfer System:Operational Principle, Model, And S- imulation of the Influence of Turbulence and Noise [J].Optics Communications, 2012,285(6):905-928.
[68]Sh. J. Zhuo, T. Sim. Defocus Map Estimation from a Single Image [J]. Pattern Recognition,2011,44(9):1852-1858.
[69]A. N. J. Raj, R. C. Staunton. Rational Filter Design for Depth from Defocus [J]. Pattern Recognition,2012,45(1):198-207.
[70]S. Sebastian, J. Burge, W. Geisler. Human Defocus Blur Discrimination in Natural Images [J]. Journal of Vision,2012,12(9):279-279.
[71]X. N. Cui, Zh. Y. Shi, J. N. Lin, L. F. Huang. The Research of Image Quality Assessment Methods [J]. Physics Procedia,2012,25:485-491.
[72]M. Carnec, P. L. Callet, D. Barba. Objective Quality Assessment of Color Images Based on a Generic Perceptual Reduced Reference [J]. Signal Processing:Image Communication,2008,23(4):239-256.
[73]L. Ma, S. N. Li, K. N. Ngan. Reduced-reference Image Quality Assessment in Reorganized DCT Domain [J]. Signal Processing:Image Communication, In Press, and Available online,2012.
[74]U. Engelke, A. Maeder, H. J. Zepernick. Human Observer Confidence in Image Quality Assessment [J]. Signal Processing:Image Communication,2012,27(9): 935-947.
[75]W. Lu, K. Zeng, D. Ch. Tao, Y. Yuan, X. B. Gao. No-reference Image Quality Assessment in Contourlet Domain [J]. Neurocomputing,2010,73(4):784-794.
[76]M. G. Martini, Ch. T. E. R. Hewage, B. Villarini. Image Quality Assessment Based on Edge Preservation [J]. Signal Processing:Image Communication,2012, 27(8):875-882.
[77]J. F. Zhao, H. J. Feng, Zh. H. Xu, Q. Li. An Image Quality Assessment Based on Regional Weight [J]. Optik-International Journal for Light and Electron Optics, 2012,123(6):494-497.
[78]S. D. Chen. A New Image Quality Measure for Assessment of Histogram Equalization-based Contrast Enhancement Techniques [J]. Digital Signal Proc-essing,2012,22(4):640-647.
[79]龚洪涛.基于HVS特性的图像质量客观评价[D].南京：南京理工大学,2011.
[80]Zh. Wang, C. B. Alan. Image Quality Assessment:From Error Visibility to Str-uctural Similarity [J]. IEEE Transaction on Image Processing,2004,13(4):1-14.
[81]X. Q. Zhao, X. M. Wang. A Novel Adaptive High-performance and Nonlinear Filtering Algorithm for Mixed Noise Removal[J]. Journal of Electronic Imaging, 2012,21(2):1-15.
[82]X. Q. Zhao, X. M. Wang, Q. Liu. A High-Performance Filter for Image Processing Based on Local Features [C]. In Proc. of the 20113rd International Conference on Computer Design and Applications (ICCDA 2011), Xi'an, May 27-29,2011:174-179.
[83]X. Q. Zhao, X. M. Wang, Q. Liu. A High-Performance Filter for Image Processing Based on Local Features [J]. Applied Mechanics and Materials,2012, 182-183:496-500.
[84]Y. Y. Zhou, Zh. F. Ye, Y. Xiao. A Restoration Algorithm for Images Contamina-ted by Mixed Gaussian Plus Random-valued Impulse Noise [J]. Journal of Vis-ual Communication and Image Representation,2013,24(3):283-294.
[85]Zh. P. Wei, J. Wang, H. Nichol, S. Wiebe, D. Chapman. A Median-Gaussian Filtering Framework for Moire Pattern Noise Removal from X-ray Microscopy Image [J]. Micron,2012,43(2):170-176.
[86]Ch. T. Lu, T. Ch. Chou. Denoising of Salt-and-pepper Noise Corrupted Image Using Modified Directional-weighted-median Filter [J]. Pattern Recognition Letters,2012,33(10):1287-1295.
[87]李文祥.图像统计建模及其在图像取证中的应用研究[D].河南：解放军信息工程大学,2010.
[88]翟艳霞.基于统计模型的SAR图像分割[D].西安：西安电子科技大学,2010.
[89]L. A. Zadeh. Fuzzy sets [J]. Information and Control,1965,8(3):338-353.
[90]梁娟.基于模糊推理的噪声图像边缘检测技术的研究[D].西安：陕西师范大学,2008.
[91]刘锁兰.基于模糊理论的图像分割区域法研究[D].南京：南京理工大学,2008.
[92]P. T. Yu, Y. L. Chen, B. M. Chang. A High Performance Filter Based on Statistic Methods for Image Processing[C]. In Proc. of the 8th Intelligent Systems Design and Applications (ISDA), Kaohsiung, Nov.26-28,2008:505-510.
[93]V. B. Nikolaos, X. Ch. Zhiwei. Gait Recognition Using Radon Transform and Linear Discriminant analysis[J]. IEEE Trans. On Image Processing,2007,16(3): 731-740.
[94]甘俊英,何思斌.非线性Radon变换及其在人脸识别中的应用[J].模式识别 与人工智能,2011,24(3)：405-410.
[95]Ch. H. Lin, J. Sh. Tsai, Ch. T. Chiu. Switching Bilateral Filter with a Texture /Noise Detector for Universal Noise Removal[J]. IEEE Trans. Image Process-ing, 2010,19(9):2307-2320.
[96]J. Martti, K. Jyrki, R. Timo. Comparison of Algorithms for Standard Median Filtering[J]. IEEE Trans. Signal Processing,1991,39(1):204-208.
[97]T. A. Nguyen, W. S. Song, M. Ch. Hong. Spatially Adaptive Denoising Algori-thm for A Single Image Corrupted by Gaussian Noise[J]. IEEE Trans. Consum-er Electronics,2010,56(3):1610-1614.
[98]B. Li, Q. Sh. Liu, J. W. Xu, X. J. Luo. A New Method for Removing Mixed Noises[J]. Science China, Information Sciences,2011,54(1):51-59.
[99]J. Tian, L. Chen. Image Noise Estimation Using a Variation-Adaptive Evolutio-nary Approach [J]. IEEE Signal Processing Letters,2012,19(7):395-398.
[100]S. H. Lee, H. M. Park, S. Y. Hwang. Motion Deblurring Using Edge Map with Blurred/Noisy Image Pairs [J]. Optics Communications,2012,285(7):1777-1786.
[101]阮秋琦,仵冀颖.数字图像处理中的偏微分方程方法[J].信号处理,2012,28(3)：301-314.
[102]F. Zhang, Y. M. Yoo, L. M. Koh, Y. M. Kim. Nonlinear Diffusion in Laplacian Pyramid Domain for Ultrasonic Speckle Reduction [J]. IEEE Trans, on medical imaging,2007,26(2):200-211.
[103]L. Alvarezl, J. Esclarin. Image Quantization Using Reaction-diffusion Equation [J]. SIAM Journal on Applied Mathematics,1997,57(1):153-175.
[104]X. Q. Zhao, X. M. Wang, L. Ch. Zhang. Reaction-Diffusion Equation Based Image Denoising Algorithm[J]. Chinese Journal of Electronics,2012,21(3): 495-499.
[105]S. M. Borodkin, A. M. Borodkin, I. B. Muchnik. Optimal Requantization of Deep Grayscale Images and Lloyd-Max Quantization[J]. IEEE Trans. On Image Process,2006,15(2):445-448.
[106]周雪媛.基于偏微分的图像修复算法研究[D].成都：成都理工大学,2011.
[107]Y. Chen, B. C. Vemuri, L. Wang. Image Denoising and Segmentation via Nonlinear Diffusion[J]. Computers and Mathematics with Applications,2000,39: 131-149.
[108]K. Dabov, A. Foi, V. Katkovnik, K. Egiazarian. Image Restoration by Sparse 3D Transform-domain Collaborative Filtering[C]. In Proc. of the SPIE Electr-onic Imaging, San Jose, California, January,2008(6812).
[109]A. Danielyan, V. Katkovnik, K. Egiazarian. BM3D Frames and Variational Image Deblurring[J]. IEEE Trans. On Image Process,2011,21(4):1715-172.
[110]R. Neelamani, H. Choi, R. G. Baraniuk. Forward:Fourier-wavelet Regularized Deconvolution for Ill-conditioned Systems [J]. IEEE Trans. On Image Process. 2004,52(2):418-433.
[111]K. Dabov, A. Foi, V. Katkovnik, K. Egiazarian. BM3D Image Denoising with Shape-adaptive Principal Component Analysis[C]. In Proc. of the Workshop on Signal Processing with Adaptive Sparse Structured Representations (SPARS'09), Saint-Malo, France, April,2009.
[112]李兵,刘全升,徐家伟,罗晓军.去除混合噪音的一种新方法[J].中国科学：信息科学,2010,40(9)：1165-1175.
[113]于冬岩.基于Trace变换不变特征的图像检索方法研究[D].吉林：吉林大学,2009.
[114]Z.X. Ji, Y. Xia, Q.S. Sun, D.Sh. Xia, D.D. Feng. Local Gaussian Distribution Fitting Based FCM Algorithm for Brain MR Image Segmentation [J].Intelligent Science and Intelligent Data Engineering,2012,7202:318-325.
[115]S. Lefkimmiatis, A. Bourquard, M. Unser. Hessian-based Norm Regularization for Image Restoration with Biomedical Applications [J]. IEEE Trans, on Image Processing,2012,21(3):983-995.
[116]L. Zhang, L. Zhang, D. Zhang. A Multi-scale Bilateral Structure Tensor Based Corner Detector [J]. Computer Vision-ACCV 2009,2010:618-627.
[117]T. Brox, J. Weickert, B. Bergeth, P. Mrazek. Nonlinear Structure Tensors [J]. Image and Vision Computing,2006,24(1):41-55.
[118]刘红毅.结构保持的图像去噪方法研究[D].南京：南京理工大学,2011.
[119]M. Ravudu, V. Jain, M. M. R. Kunda. Review of Image Processing Techniques for Automatic Detection of Eye Diseases[C]. In Proc. of the 2012 Sixth Inter-national Conference on Sensing Technology (ICST), Kolkata, Dec.18-21,2012: 320-325.
[120]M. Keating, D. Flynn, R. Aitken, et al. Low Power Methodology Manual:for S-ystem-on-Chip Design [M]. Springer Publishing Company, Incorporated,2007.
[121]I. F. Akyildiz, T. Melodia, R. Ch. Kaushik. A Survey on Wireless Multimedia Sensor Networks [J]. Computer Networks,2007,51 (4):921-960.
[122]A.N. Belbachir, M. Litzenberger, C. Posch and P. Schon. Real-time vision using a smart sensor system [C]. In Proc. of the International Symposium on Industrial Electronics (ISIE), Spain, June 2007:1968-1973.
[2]孙即祥.图像处理[M].北京：科学出版社,2004.
[3]G. A. Baxes. Digital Image Processing:Principles and Applications [M]. New York,1994:20-25.
[4]X. Wang, X. Q. Zhao, F. X. Guo, J. F. Ma. Impulsive Noise Detection by Double Noise Detector and Removal using Adaptive Neural-fuzzy Inference System [J]. Int. J. Electron.Commun(AEu),2011,65(5):429-434.
[5]A. K. Katsaggelos, M. R. Schroeder, T. Kohonen, et al. Digital image restora-tion [M]. Springer-Verlag New York Inc.,1991.
[6]T. S. Cho, C. L. Zitnick, N. Joshi, et al. Image Restoration by Matching Gradient Distributions [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(4):683-694.
[7]冈萨雷斯.数字图像处理(第二版)[M].北京：电子工业出版社,2003.
[8]A. Kour, V. K. Yadav, V. Maheshwari. A Review on Image Processing [J]. International Journal of Electronics Communication and Computer Engineering, 2013,4(1):270-275.
[9]M. Egmont-Petersen, D. de Ridder, H. Handels. Image Processing with Neural Networks-A Review [J]. Pattern Recognition,2002,35(10):2279-2301.
[10]J. Mairal, F. R. Bach, J. Ponce, et al. Non-local Sparse Models for Image Restoration [C]. In Proc. of the 2009 IEEE 12th International Conference on Computer Vision, Kyoto, Sept.29-Oct.2,2009:2272-2279.
[11]T. Ch. Lin. Decision-based Fuzzy Image Restoration for Noise Reduction Based on Evidence Theory [J]. Expert Systems with Applications,2011,38(7): 8303-8310.
[12]M. Bertalmio, G. Sapiro, V. Caselles, et al. Image Inpainting [C]. In Proc. of the International Conference on Computer Graphics and Interactive Techniques, Louisiana, USA,2000:417-424.
[13]G. Xu, E. Tang. Blind Deconvolution for Image Restoration Based on Text Characteristic[C]. In Proc. of the Control,2011 International Conference on Automation and Systems Engineering (CASE), Singapore, July 30-31,2011:1-3.
[14]M R Banham, A K Katsaggelos. Digital image restoration [J]. IEEE Signal Pro-cessing Magazine,1997,14(2):24-41.
[15]Computer Vision Research Groups of Carnegie Mellon University [EB/OL]. H-ttp://www.cs.cmu. edu/-cil/vision.html,2013-03-12/2013-03-12.
[16]Computer Vision and Robotics Group of Cambridge University [EB/OL]. http:/ /mi.eng.cam.ac.uk/mi/,2013-03-12/2013-03-12.
[17]Vision and Image Analysis Group of Cornell University [EB/OL]. http://ww-w.via.cornell.edu/,2013-03-12/2013-03-12.
[18]Harvard Robotics Laboratory [EB/OL]. http://hrl.harvard.edu/,2013-03-12/ 2013-03-12.
[19]Zucker Computational Vision Group [EB/OL]. http://cs-www.cs.yale.edu/hom-es/vision/zucker/,2013-03-12/2013-03-12.
[20]Stanford Vision Laboratory [EB/OL]. http://vision.stanford.edu/,2013-03-12/ 2013-03-12.
[21]Biometerics Research Group of Michican State University [EB/OL]. http://w-ww.cse.msu.edu/biometrics/,2013-03-12/2013-03-12.
[22]Computer Vision Group located within the Division Ⅲ of the Computer Science Department in the University of Bonn [EB/OL]. http.//www-dbv.informati-k.uni-bonn.de/,2013-03-12/2013-03-12.
[23]Queen Mary Vision Laboratory [EB/OL]. http://www.dcs.qmul.ac.uk/research/ vision/,2013-03-12/2013-03-12.
[24]Technical Committee on Pattern Analysis and Machine Intelligence (TCPAMI) [EB/OL]. http://www.computer.org/portal/web/tcpami,2013-03-12/2013-03-12.
[25]IPDL世界知识产权组织电子图书馆[EB/OL]. http://www.wipo.int/wipogold/ en/,2013-03-01/2013-03-01.
[26]中国科学院沈阳自动化研究所机器人视觉研究组[EB/OL]. http://vision. sia.cn/visiongroup%28chinese%29.html,2013-03-12/2013-03-12.
[27]华中科技大学图像识别与人工智能研究所[EB/OL]. http://iprai.hust.edu.cn/, 2013-03-12/2013-03-12.
[28]东南大学影像科学与技术实验室[EB/OL]. http://www.imagetech.com.cn/, 2013-03-12/2013-03-12.
[29]浙江大学应用数学与图像处理研究所[EB/OL]. http://www.cms.zju.edu.cn/ frontindex.asp,2013-03-12/2013-03-12.
[30]国家自然科学基金委员会资助项目统计[EB/OL]. http://www.nsfc.gov. cn/nsfc/cen/xmtj/index.html,2013-03-12/2013-03-12.
[31]沈垣,李舜酩,毛建国,辛江慧.数字图像复原技术综述[J].中国图象图形学报,2009,14(9)：1764-1775.
[32]E. O. Brigham, H. W. Smith, F. X. Bostick, et al. An Iterative Technique for Determining Inverse Filters [J]. IEEE Trans, on Geoscience Electronics,1968, 6(2):86-96.
[33]N. Mastronardi, P. Lemmerling, A. Kalsi, et al. Implementation of the Regularized Structured Total Least Squares Algorithms for Blind Image Deblurring [J]. Linear Algebra and its Applications,2004,391(1):203-221.
[34]B. Chalmond. Image Restoration Using an Estimated Markov Model [J]. Sign-al Processing 1988,15(2):115-129.
[35]M. I. Sezan, A. M.T ekalp. Adaptive Nonlinear Image Restoration Using the Theory of Convex Projections [J]. Circuits and Systems,1988,7(9):959-962.
[36]Jr. N. C. Gallagher, G. L. Wise. A Theoretical Analysis of Properties of the Median Filters [J]. IEEE Trans. On Acoustics Speech, Signal Processing,1981, 29(1):1136-1141.
[37]J. Serra. Morphological Filtering:An overview [J]. Signal Processing,1994,38 (1):3-11.
[38]F. Farbiz, M. Menhaj, S. Motamedi, and M. Hagan. A New Fuzzy Logic Filter for Image Enhancement [J]. IEEE Trans. Syst, Man, Cybern. B, Cybern.2000, 30(1):110-119.
[39]P. Windyga. Fast Impulsive Noise Removal [J]. IEEE Transaction on Image Process.2001,10(1):173-179.
[40]Y. T. T. Zhou, R. Chellappa, A. Vaid. Image Restoration Using a Neural Net-work [J]. IEEE Transactions on Acoustics, Speech and Signal Processing,1988, 36(7):1141-1151.
[41]Y. D. Wu, Y. H. Chen, H. Y. Zhang. An Improved Algorithm for Image Restora-tion based on Modified Hopfield Neural Network [J]. Machine Learning and Cybernetics,2005,48(7):4720-4723.
[42]L. J. Deng, T. Zh. Huang, X. L. Zhao. Wavelet-based Two-level Methods for Image Restoration [J]. Communications in Nonlinear Science and Numerical Simulation,2012,17(12):5079-5087.
[43]L. P. Zhang, H. Y. Zhang, H. F. Shen, P. X. Li. A Super-resolution Reconstruction Algorithm for Surveillance Images [J]. Signal Processing,2010,90(3):848-859.
[44]Ch. P. Sung, K. P. Min, G K. Moon. Super-resolution Image Reconstruction:a Technical Overview [J]. IEEE Signal Processing Magazine,2003,20(3):21-36.
[45]X. F. Wan, Y. Yang. Super-resolution Image Reconstruction[C]. In Proc. of the 2010 International Conference on Computer Application and System Modeling (ICCASM), Taiyuan, Oct.22-24,2010(8):351-355.
[46]K. K. Delibasis, P. E. Undrill, G G Cameron. Genetic Algorithm Implement-ation of Stack Filter Design for Image Restoration [J]. IEE Proceedings Vision, Image and Signal Processing,1996,143(3):177-183.
[47]E. Y. Kim, K. T. Kim, B. Kim. Genetic Algorithm-based Reconstruction of Old Films Corrupted by Scratches and Blotches [J]. Pattern Recognition Letters,2013, 34(2):226-237.
[48]O. P. Verma, P. Kumar, M. Hanmandlu, S. Chhabra. High Dynamic Range Optimal Fuzzy Color Image Enhancement Using Artificial Ant Colony System [J]. Applied Soft Computing,2012,12(1):394-404.
[49]A. M. A. C. Rocha, M. F. P. Costa. Artificial Fish Swarm Filter-Based Method for Constrained Global Optimization [J]. Computational Science and Its Appli-cations-ICCSA,2012,7335:57-71.
[50]A. Buades, B. Coll, and J. M. Morel. A Review of Image Denoising Algorithms, with a New One [J]. Multiscale Model Simul.,2006,4(2),490-530.
[51]徐树奎.基于计算摄影的运动模糊图像清晰化技术研究[D].湖南：国防科技大学,2011.
[52]J. F. Cai, B. Dong, S. Osher, Z. W. Shen. Image Restoration:Total Variation, W-avelet Frames and Beyond [J]. J. Amer. Math. Soc,2012,25:1033-1089.
[53]S. Zheng, Z. Pan, G Wang, X. Y. Audio. A Variational Model of Image Restoration Based on First and Second Order Derivatives and Its Split Bregman algorithm[C]. In Proc. of the 2012 International Conference on Audio, Language and Image Processing (ICALIP), Shanghai, July 16-18,2012:860-865.
[54]C. Liu, W. T. Freeman, R. S. Szeliski, et al. Noise Estimation from a Single Im-age [C]. In Proc. of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 17-22,2006,1:901-908.
[55]Y. Hwang J. S. Kim, I. S. Kweon. Difference-Based Image Noise Modeling Us-ing Skellam Distribution [J]. IEEE Transaction on Pattern Analysis and Machi-ne Intelligence,2012,34(7):1329-1341.
[56]Y. Y. Zhou, Zh. F. Ye, Y. Xiao. A Restoration Algorithm for Images Contamina-ted by Mixed Gaussian Plus Random-valued Impulse Noise [J]. Journal of Vis-ual Communication and Image Representation,2013,24(3):283-294.
[57]Zh. P. Wei, J. Wang, H. Nichol, S. Wiebe, D. Chapman. A Median-Gaussian Fi-ltering Framework for Moire Pattern Noise Removal from X-ray Microscopy I-mage [J]. Micron,2012,43(2-3):170-176.
[58]Ch. T. Lu, T. Ch. Chou. De-noising of Salt-and-pepper Noise Corrupted Image Using Modified Directional-weighted-median Filter [J]. Pattern Recognition L-etters,2012,33(10):1287-1295.
[59]Y. Han, X. Ch. Feng, G. Baciu, W. W. Wang. Nonconvex Sparse Regularizer B-ased Speckle Noise Removal [J]. Pattern Recognition,2013,46(3):989-1001.
[60]邹谋炎.反卷积和信号复原[M].北京：国防工业出版社,2001.
[61]刘益民.图像运动模糊机理与恢复技术方法的研究[D].南京：南京理工大学,2008.
[62]刘明.基于图像复原的航空摄影前向像移检测及补偿技术研究[D].长春：中国科学院研究生院(长春光学精密机械与物理研究所),2005.
[63]M. Yang, Ch. L. Gong. Underwater Image Restoration by Turbulence Model Based on Image Gradient Distribution[C]. In Proc. of the 2nd International Co-nference on Uncertainty Reasoning and Knowledge Engineering (URKE), Jal-arta, Aug 14-15,2012:296-299.
[64]李庆菲.大气湍流退化图像复原研究[D].安徽：合肥工业大学,2010.
[65]D. F. Shi, Ch. Y. Fan, H. Shen, P. F. Zhang, et al. Restoration of Atmospheric T-urbulence Degraded Images [J]. Optics Communications,2011,284(24):5556-5561.
[66]H. Y. Hong, L. Ch. Li, T. X. Zhang. Blind Restoration of Real Turbulence-degr aded Image with Complicated Backgrounds Using Anisotropic Regularization [J]. Optics Communications,2012,285(24):4977-4986.
[67]V. P. Aksenov, I. V. Izmailov, F. Y. Kanev, B. N. Poizner. Optical Vortex Detect-or as A Basis for A Data Transfer System:Operational Principle, Model, And S- imulation of the Influence of Turbulence and Noise [J].Optics Communications, 2012,285(6):905-928.
[68]Sh. J. Zhuo, T. Sim. Defocus Map Estimation from a Single Image [J]. Pattern Recognition,2011,44(9):1852-1858.
[69]A. N. J. Raj, R. C. Staunton. Rational Filter Design for Depth from Defocus [J]. Pattern Recognition,2012,45(1):198-207.
[70]S. Sebastian, J. Burge, W. Geisler. Human Defocus Blur Discrimination in Natural Images [J]. Journal of Vision,2012,12(9):279-279.
[71]X. N. Cui, Zh. Y. Shi, J. N. Lin, L. F. Huang. The Research of Image Quality Assessment Methods [J]. Physics Procedia,2012,25:485-491.
[72]M. Carnec, P. L. Callet, D. Barba. Objective Quality Assessment of Color Images Based on a Generic Perceptual Reduced Reference [J]. Signal Processing:Image Communication,2008,23(4):239-256.
[73]L. Ma, S. N. Li, K. N. Ngan. Reduced-reference Image Quality Assessment in Reorganized DCT Domain [J]. Signal Processing:Image Communication, In Press, and Available online,2012.
[74]U. Engelke, A. Maeder, H. J. Zepernick. Human Observer Confidence in Image Quality Assessment [J]. Signal Processing:Image Communication,2012,27(9): 935-947.
[75]W. Lu, K. Zeng, D. Ch. Tao, Y. Yuan, X. B. Gao. No-reference Image Quality Assessment in Contourlet Domain [J]. Neurocomputing,2010,73(4):784-794.
[76]M. G. Martini, Ch. T. E. R. Hewage, B. Villarini. Image Quality Assessment Based on Edge Preservation [J]. Signal Processing:Image Communication,2012, 27(8):875-882.
[77]J. F. Zhao, H. J. Feng, Zh. H. Xu, Q. Li. An Image Quality Assessment Based on Regional Weight [J]. Optik-International Journal for Light and Electron Optics, 2012,123(6):494-497.
[78]S. D. Chen. A New Image Quality Measure for Assessment of Histogram Equalization-based Contrast Enhancement Techniques [J]. Digital Signal Proc-essing,2012,22(4):640-647.
[79]龚洪涛.基于HVS特性的图像质量客观评价[D].南京：南京理工大学,2011.
[80]Zh. Wang, C. B. Alan. Image Quality Assessment:From Error Visibility to Str-uctural Similarity [J]. IEEE Transaction on Image Processing,2004,13(4):1-14.
[81]X. Q. Zhao, X. M. Wang. A Novel Adaptive High-performance and Nonlinear Filtering Algorithm for Mixed Noise Removal[J]. Journal of Electronic Imaging, 2012,21(2):1-15.
[82]X. Q. Zhao, X. M. Wang, Q. Liu. A High-Performance Filter for Image Processing Based on Local Features [C]. In Proc. of the 20113rd International Conference on Computer Design and Applications (ICCDA 2011), Xi'an, May 27-29,2011:174-179.
[83]X. Q. Zhao, X. M. Wang, Q. Liu. A High-Performance Filter for Image Processing Based on Local Features [J]. Applied Mechanics and Materials,2012, 182-183:496-500.
[84]Y. Y. Zhou, Zh. F. Ye, Y. Xiao. A Restoration Algorithm for Images Contamina-ted by Mixed Gaussian Plus Random-valued Impulse Noise [J]. Journal of Vis-ual Communication and Image Representation,2013,24(3):283-294.
[85]Zh. P. Wei, J. Wang, H. Nichol, S. Wiebe, D. Chapman. A Median-Gaussian Filtering Framework for Moire Pattern Noise Removal from X-ray Microscopy Image [J]. Micron,2012,43(2):170-176.
[86]Ch. T. Lu, T. Ch. Chou. Denoising of Salt-and-pepper Noise Corrupted Image Using Modified Directional-weighted-median Filter [J]. Pattern Recognition Letters,2012,33(10):1287-1295.
[87]李文祥.图像统计建模及其在图像取证中的应用研究[D].河南：解放军信息工程大学,2010.
[88]翟艳霞.基于统计模型的SAR图像分割[D].西安：西安电子科技大学,2010.
[89]L. A. Zadeh. Fuzzy sets [J]. Information and Control,1965,8(3):338-353.
[90]梁娟.基于模糊推理的噪声图像边缘检测技术的研究[D].西安：陕西师范大学,2008.
[91]刘锁兰.基于模糊理论的图像分割区域法研究[D].南京：南京理工大学,2008.
[92]P. T. Yu, Y. L. Chen, B. M. Chang. A High Performance Filter Based on Statistic Methods for Image Processing[C]. In Proc. of the 8th Intelligent Systems Design and Applications (ISDA), Kaohsiung, Nov.26-28,2008:505-510.
[93]V. B. Nikolaos, X. Ch. Zhiwei. Gait Recognition Using Radon Transform and Linear Discriminant analysis[J]. IEEE Trans. On Image Processing,2007,16(3): 731-740.
[94]甘俊英,何思斌.非线性Radon变换及其在人脸识别中的应用[J].模式识别 与人工智能,2011,24(3)：405-410.
[95]Ch. H. Lin, J. Sh. Tsai, Ch. T. Chiu. Switching Bilateral Filter with a Texture /Noise Detector for Universal Noise Removal[J]. IEEE Trans. Image Process-ing, 2010,19(9):2307-2320.
[96]J. Martti, K. Jyrki, R. Timo. Comparison of Algorithms for Standard Median Filtering[J]. IEEE Trans. Signal Processing,1991,39(1):204-208.
[97]T. A. Nguyen, W. S. Song, M. Ch. Hong. Spatially Adaptive Denoising Algori-thm for A Single Image Corrupted by Gaussian Noise[J]. IEEE Trans. Consum-er Electronics,2010,56(3):1610-1614.
[98]B. Li, Q. Sh. Liu, J. W. Xu, X. J. Luo. A New Method for Removing Mixed Noises[J]. Science China, Information Sciences,2011,54(1):51-59.
[99]J. Tian, L. Chen. Image Noise Estimation Using a Variation-Adaptive Evolutio-nary Approach [J]. IEEE Signal Processing Letters,2012,19(7):395-398.
[100]S. H. Lee, H. M. Park, S. Y. Hwang. Motion Deblurring Using Edge Map with Blurred/Noisy Image Pairs [J]. Optics Communications,2012,285(7):1777-1786.
[101]阮秋琦,仵冀颖.数字图像处理中的偏微分方程方法[J].信号处理,2012,28(3)：301-314.
[102]F. Zhang, Y. M. Yoo, L. M. Koh, Y. M. Kim. Nonlinear Diffusion in Laplacian Pyramid Domain for Ultrasonic Speckle Reduction [J]. IEEE Trans, on medical imaging,2007,26(2):200-211.
[103]L. Alvarezl, J. Esclarin. Image Quantization Using Reaction-diffusion Equation [J]. SIAM Journal on Applied Mathematics,1997,57(1):153-175.
[104]X. Q. Zhao, X. M. Wang, L. Ch. Zhang. Reaction-Diffusion Equation Based Image Denoising Algorithm[J]. Chinese Journal of Electronics,2012,21(3): 495-499.
[105]S. M. Borodkin, A. M. Borodkin, I. B. Muchnik. Optimal Requantization of Deep Grayscale Images and Lloyd-Max Quantization[J]. IEEE Trans. On Image Process,2006,15(2):445-448.
[106]周雪媛.基于偏微分的图像修复算法研究[D].成都：成都理工大学,2011.
[107]Y. Chen, B. C. Vemuri, L. Wang. Image Denoising and Segmentation via Nonlinear Diffusion[J]. Computers and Mathematics with Applications,2000,39: 131-149.
[108]K. Dabov, A. Foi, V. Katkovnik, K. Egiazarian. Image Restoration by Sparse 3D Transform-domain Collaborative Filtering[C]. In Proc. of the SPIE Electr-onic Imaging, San Jose, California, January,2008(6812).
[109]A. Danielyan, V. Katkovnik, K. Egiazarian. BM3D Frames and Variational Image Deblurring[J]. IEEE Trans. On Image Process,2011,21(4):1715-172.
[110]R. Neelamani, H. Choi, R. G. Baraniuk. Forward:Fourier-wavelet Regularized Deconvolution for Ill-conditioned Systems [J]. IEEE Trans. On Image Process. 2004,52(2):418-433.
[111]K. Dabov, A. Foi, V. Katkovnik, K. Egiazarian. BM3D Image Denoising with Shape-adaptive Principal Component Analysis[C]. In Proc. of the Workshop on Signal Processing with Adaptive Sparse Structured Representations (SPARS'09), Saint-Malo, France, April,2009.
[112]李兵,刘全升,徐家伟,罗晓军.去除混合噪音的一种新方法[J].中国科学：信息科学,2010,40(9)：1165-1175.
[113]于冬岩.基于Trace变换不变特征的图像检索方法研究[D].吉林：吉林大学,2009.
[114]Z.X. Ji, Y. Xia, Q.S. Sun, D.Sh. Xia, D.D. Feng. Local Gaussian Distribution Fitting Based FCM Algorithm for Brain MR Image Segmentation [J].Intelligent Science and Intelligent Data Engineering,2012,7202:318-325.
[115]S. Lefkimmiatis, A. Bourquard, M. Unser. Hessian-based Norm Regularization for Image Restoration with Biomedical Applications [J]. IEEE Trans, on Image Processing,2012,21(3):983-995.
[116]L. Zhang, L. Zhang, D. Zhang. A Multi-scale Bilateral Structure Tensor Based Corner Detector [J]. Computer Vision-ACCV 2009,2010:618-627.
[117]T. Brox, J. Weickert, B. Bergeth, P. Mrazek. Nonlinear Structure Tensors [J]. Image and Vision Computing,2006,24(1):41-55.
[118]刘红毅.结构保持的图像去噪方法研究[D].南京：南京理工大学,2011.
[119]M. Ravudu, V. Jain, M. M. R. Kunda. Review of Image Processing Techniques for Automatic Detection of Eye Diseases[C]. In Proc. of the 2012 Sixth Inter-national Conference on Sensing Technology (ICST), Kolkata, Dec.18-21,2012: 320-325.
[120]M. Keating, D. Flynn, R. Aitken, et al. Low Power Methodology Manual:for S-ystem-on-Chip Design [M]. Springer Publishing Company, Incorporated,2007.
[121]I. F. Akyildiz, T. Melodia, R. Ch. Kaushik. A Survey on Wireless Multimedia Sensor Networks [J]. Computer Networks,2007,51 (4):921-960.
[122]A.N. Belbachir, M. Litzenberger, C. Posch and P. Schon. Real-time vision using a smart sensor system [C]. In Proc. of the International Symposium on Industrial Electronics (ISIE), Spain, June 2007:1968-1973.